BSEC 2.4 + BME688 to detect BSEC_OUTPUT_GAS_ESTIMATE_1

[jack3]

Hello,
I am using the bsec_serialized_configurations_selectivity configure string as well as BSEC_SAMPLE_RATE_SCAN, however I keep getting unexpected values for GAS_ESTIMATE1 and GAS_ESTIMATE2, that value being equal to zero and GAS_ESTIMATE_x equal to zero. This value never changes no matter the environment inserted?

bsec_serialized_configurations_selectivity are from BME AI-Studio-v2.0.0-win

Thanks a lot!

[BSTRobin]

Hi jack3,

You need to collect data for the gases you need to detect, import the data into BME AI-Studio, generate the configuration file you want, and replace your configuration file before running BSEC2.4.0.0.

[jack3]

Hi BSTRobin,

I collected gas data using the BME688 development kit and created an algorithm using BME AI Studio v2.0.0-win, and successfully exported the BSEC configuration file.

Using the BME688 development kit, normal air was detected in the gas algorithm files (. aiconfig,. config). Afterwards, I changed the project configuration files (.c,.h), but did not detect normal air, I kept getting GAS_ ESTIMATE_ 1 and GAS_ ESTMATE_2 Unexpected value, which is equal to zero, GAS_ ESTIMATE_ X (Accuracy Value) is equal to zero.

Does the "*.config" file have any role in the BSEC 2.4 configuration? I didn't use it.

[BSTRobin]

Hi jack3,

There is a ±15 % deviation between sensors, which may lead to differences in gas recognition results.
Latest BSEC v2.5 and BME AI Studio v2.3.2 support optimization for BME688 gas detection, resulting in smaller differences between different sensors running the same configuration.